tl;dr: Xiaomi-Robotics-1 is a robot foundation model pre-trained on 100,000 hours of embodiment-free UMI trajectories, then post-trained on ~7,200 hours of real-robot data to align with physical embodiments and natural-language instructions. The team reports clean scaling laws—larger models and more data steadily improve both validation error and real-robot success rates—and shows the model learns new tasks like phone packing and laundry loading from under 10 hours of demos, roughly doubling π0.5's success rate. It also achieves state-of-the-art results on four simulation benchmarks including RoboCasa and VLABench.